This invention relates to signal conditioning circuits, which can be used to condition the output of trigger probes. Trigger probes are used for measuring workpieces on position determination apparatus such as coordinate measuring machines or machine tools, and deliver a trigger signal when a position measurement is to be taken.
A known type of such probe is described in U.S. Pat. No. 4,153,998 (McMurtry). It comprises a workpiece-contacting stylus which is biased by a spring into a rest position defined by kinematically arranged seating elements. A normally closed electrical circuit through the kinematic seating elements is opened when the stylus is deflected by contact with a workpiece, and the opening of this electrical circuit is detected to provide a trigger signal indicating contact between the stylus and the workpiece.
In practice, such probes are sold with an interface circuit which monitors the resistance across the kinematic seating elements. The trigger signal is developed at the output of the interface when the resistance across the contacts formed by the seating elements increases above a certain predetermined threshold, as the contacts open. The output signal from the interface reverts to its original state to indicate when the contacts re-close, i.e. when the stylus has correctly reseated in its kinematic seating elements and the resistance across the contacts falls.
The kinematic seating arrangement of the probe ensures that the stylus always returns to an extremely repeatable mechanical rest position, and the interface circuit discussed above ensures that the electrical generation of the trigger signal is also very repeatable. This repeatability ensures that the probe gives highly accurate results when used on position determination apparatus.
Very occasionally, however, a problem is experienced after a probe has been in use for a long time. After the stylus has returned to its rest position, the output of the interface still indicates that the stylus has not correctly reseated in the kinematic seating arrangement. Such a problem has to be rectified before further measurements can be taken. This is especially inconvenient if the probe is being used to take a sequence of measurements automatically, as is nowadays usually the case.
Our investigations of this problem show that the mechanical reseating of the stylus in its kinematically-defined rest position is in general perfectly satisfactory. We have now found that the problem arises, particularly in a probe which has been used for some time, because the resistance across the seating elements when mechanically reseated is no longer zero, but can rise above the threshold which is detected by the interface. The interface is then unable to distinguish between the seated and unseated states of the stylus.